Human in Events

Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve)


Short Name: LightTrack (Baseline)
Complete name: LightTrack (Baseline)
Description: LightTrack (Baseline)
Runtime: none Hz
Submissions: 1

Benchmark
performance:

MOTA MOTP Total_AP
27.2089 55.1711 29.2119

Detailed
performance:


Video Name MOTA MOTP Total_AP
hm_in_waiting_hall (ID:20) 32.2670 58.8612 35.7083
hm_in_bus (ID:21) 10.4107 27.1695 7.4001
hm_in_dining_room2 (ID:22) 11.8139 24.1271 11.9560
hm_in_lab2 (ID:23) 41.0784 54.4799 48.0989
hm_in_subway_station (ID:24) 18.1635 39.3365 20.7105
hm_in_passage (ID:25) 17.2739 58.1977 20.0364
hm_in_fighting4 (ID:26) 25.5640 62.8177 26.0566
hm_in_shopping_mall3 (ID:27) 23.0039 59.8947 23.7352
hm_in_restaurant (ID:28) 17.0871 50.0185 16.8753
hm_in_accident (ID:29) 29.6614 53.3939 29.9826
hm_in_stair3 (ID:30) 33.6290 39.0933 30.4931
hm_in_crossroad (ID:31) 42.7915 71.2257 47.1709
hm_in_robbery (ID:32) 30.1257 50.8328 31.7720